Attorney General's Public Counsel Unit, V Wa Utilities & Transportation Comm

Filed Washington State Court of Appeals Division Two

August 7, 2018

IN THE COURT OF APPEALS OF THE STATE OF WASHINGTON

DIVISION II WASHINGTON ATTORNEY GENERAL’S No. 48982-1-II OFFICE, PUBLIC COUNSEL UNIT,

Appellant,

v.

WASHINGTON UTILITIES AND PUBLISHED OPINION TRANSPORATION COMMISSION, a Washington State agency,

Respondent,

AVISTA CORPORATION d/b/a AVISTA UTILITIES,

Intervenor/Respondent.

MELNICK, J. — The Washington Attorney General’s Office, Public Counsel Unit (PCU), appeals from the final order of the Washington Utilities and Transportation Commission (WUTC) setting Avista Corporation’s 2016 rates for electric and natural gas services.

We conclude that the WUTC’s use of an attrition adjustment to calculate Avista’s rate base violated RCW 80.04.250. Accordingly, we reverse and remand to the WUTC for recalculation of Avista’s electric and natural gas rates.

FACTS

Avista is an electric and natural gas company that serves customers in Washington. The WUTC has the statutory authority to regulate rates of utility companies such as Avista. RCW 80.01.040. *2 48982-1-II

Avista filed a request with the WUTC to increase its rates for electric service by $33.2 million (6.7 percent) and its natural gas service by $12 million (6.9 percent). Avista based its proposed rate increase in part on an attrition adjustment for both its electric and natural gas operations. The WUTC ultimately set rates significantly lower than Avista’s initial request, basing its analysis on an attrition study and attrition adjustment. Whether to apply this attrition adjustment in setting Avista’s rates is the fundamental dispute in this case.

I. THE PARTIES

After Avista petitioned to change its rates, WUTC Staff (Staff) and PCU both entered notice of their appearance in the subsequent joint ratemaking case. 1 The Industrial Customers of Northwest Utilities (ICNU), the Northwest Industrial Gas Users (NWIGU), and The Energy Project all intervened in the case without opposition.

II. UTILITY RATEMAKING BACKGROUND

A. RATEMAKING PROCESS

People’s Organization for Washington Energy Resources v. Utilities & Transportation Commission discussed the background principles of ratemaking in Washington. 104 Wn.2d 798, 808-11, 711 P.2d 319 (1985) (POWER 85). The court explained:

In this state, the Legislature has conferred the ratemaking power on the WUTC, subject, of course, to appropriate judicial review. Most states delegate their ratemaking power to regulatory agencies in very broad terms, basically just directing them to set those rates which the agencies determine to be just and reasonable. Washington is such a state. . . . The statutory mandate to the WUTC is to set fair, reasonable and sufficient rates. Indeed, as this court has also observed, the paramount objective of the Legislature in creating the commission, now the WUTC, “was to secure for the public safe, adequate, and sufficient utility services at just, fair, reasonable, and sufficient rates.”

1 WUTC Staff participate as a party in ratemaking proceedings. See, e.g., PacifiCorp v. Wash. Utils. & Transp. Comm’n, 194 Wn. App. 571, 578 n.2, 376 P.3d 389 (2016). PCU entered its appearance as representative of the people of the State of Washington. RCW 80.01.100.

48982-1-II POWER 85, 104 Wn.2d at 808 (footnotes omitted) (citations omitted) (quoting State ex rel. PUD 1 v. Dep’t of Pub. Serv., 21 Wn.2d 201, 209, 150 P.2d 709 (1944)). It also outlined the formula the WUTC used to set rates:

In order to control aggregate revenue and set maximum rates, regulatory commissions such as the WUTC commonly use and apply the following equation:

R = [O] + B(r)

In this equation, R is the utility’s allowed revenue requirements; O is its operating expenses; B is its rate base; and r is the rate of return allowed on its rate base.

POWER 85, 104 Wn.2d at 808-09. Rate base “represents the total investment in, or fair value of, the facilities of the utility employed in providing its service. Calculation of the rate base is of obvious importance since the product of the rate base (B) multiplied by the allowed rate of return (r) accrues to the utility’s investors.” POWER 85, 104 Wn.2d at 809 (footnote omitted).

No matter how rates are determined, they must “‘enable the company to operate successfully, to maintain its financial integrity, to attract capital, and to compensate its investors for the risks assumed.” POWER 85, 104 Wn.2d at 811 (quoting Fed. Power Comm’n v. Hope Nat. Gas Co., 320 U.S. 591, 605, 64 S. Ct. 281, 88 L. Ed. 333 (1944) (Hope)). A utility is entitled to rates that will permit it to earn a return on its investment in property it uses to serve the public. POWER 85, 104 Wn.2d at 813 (quoting Bluefield Water Works & Imp. Co. v. Public Serv. Comm’n, 262 U.S. 679, 692, 43 S. Ct. 675, 67 L. Ed. 1176 (1923)).

Before the present case, the WUTC’s standard practice was to use the “‘hybrid’ or ‘modified’ historical test year” approach. Administrative Record (AR) at 3819. This process *4 48982-1-II involved evaluating a historical test year2 and applying pro forma adjustments3 to estimate the utility’s rate base and operating expenses in the rate year. Pro forma plant adjustments provided the WUTC with a method of evaluating expected rate base changes between the historic model year and the rate year. The WUTC considered whether such adjustments were “‘known and measureable’ and ‘used and useful’ for service in Washington State.” AR at 3820.

When the historical test year relationship between revenues, expenses, and rate base does not hold during the rate year, erosion may “deprive the utility of a reasonable opportunity to earn a fair rate of return.” AR at 3836. This erosion of a company’s rate of return over time is generally referred to as “attrition.”

B. ATTRITION

As described by the WUTC,

[A]ttrition occurs when the test-period relationship between rate base, expenses and revenues does not hold under conditions in the rate effective period, such that a utility’s expenses or rate base grows more quickly than revenues, and a utility would likely have no reasonable opportunity to earn its allowed rate of return.

AR at 703-04.

An attrition adjustment, then, is “a discrete adjustment to the modified historical test year that the [WUTC] may use when it determines attrition is present.” AR at 704. When developing an attrition adjustment, parties first provide a revenue requirement analysis based on a modified historical test year to the WUTC. “Parties then perform an attrition study to determine the utility’s revenue requirement in the rate year. The attrition adjustment is the difference between the

2 The historical test year is generally the most recent year before the rates are being set for the following year. In this case, the WUTC considered the year beginning October 1, 2013, and ending September 30, 2014. 3 These adjustments include any major plant additions or other expected changes in rate base between the test year and the rate year.

48982-1-II revenue requirement provided by the modified historical test year and the revenue requirement provided by the attrition study.” AR at 704 n.60.

Attrition adjustments were frequently used in Washington as a WUTC ratemaking tool “‘during the early 1980’s in an environment of exceptional inflation and high interest rates.’” AR at 706 (quoting Wash. Utils. & Transp. Comm’n v. Puget Sound Energy, Inc., Dockets UE-111048 and UG-111049 (consolidated), Order 08 at 180 (May 7, 2012) (PSE Order 08)).

The two most common sources of attrition in Washington are “abnormal or excessive inflation and exceptional and prolonged levels of plant additions.” AR at 711. The last time the WUTC addressed attrition in a fully litigated rate case was in 1993. AR at 705. It denied the request and announced that attrition adjustments were only appropriate in “‘extraordinary circumstances’” when “‘without such an adjustment, the company would have no reasonable opportunity to earn its authorized rate of return.’” AR at 705 (quoting Wash. Utils. & Transp. Comm’n v. Wash. Nat. Gas, Docket UG-920840, Fourth Supp. Order at 30 (Sept. 27, 1993). Before the present case, the WUTC had not directly authorized an attrition adjustment since the 1980s. AR at 723.

In 2012, for the first time in almost twenty years, the WUTC suggested the use of an attrition adjustment in a ratemaking case for Puget Sound Energy. AR at 706. In that case, the WUTC observed that:

“‘[A]ttrition’ in the context of utility ratemaking is limited to circumstances in which key assumptions that underlie ratemaking theory fail to hold in reality. Regardless whether an historical or budgeted test-period is used, the relationship between rate base, expenses and revenues is used to represent the future and to set prospective rates adequate to allow a reasonable return. Ratemaking rests on the key assumption that the test-period relationships will accurately represent relationships in the future. If this assumption fails, cost of service may increase more rapidly than revenues and the rates approved based on test period conditions may not be adequate to achieve the allowed level of return under future conditions.”

48982-1-II AR at 706 (quoting PSE Order 08 at 180). The WUTC also discussed attrition in Avista’s last two rate-setting cases. AR at 706.

In Avista’s 2012 ratemaking case, the WUTC opined that consideration of attrition was appropriate in setting 2013 rates, despite it being “‘caused substantially by Avista’s ongoing capital investment program’” and that the WUTC “‘ha[d] no absolute assurance that Avista [would] complete the projects described in its plan for 2013.’” AR at 707 (quoting Wash. Utils. & Transp. Comm’n v Avista Corp., Dockets UE-120436 and UG-120437 (consolidated), Order 09 and Dockets UE-110876 and UG-110877 (consolidated), Order 14 at 4 (Dec. 26, 2012) (Avista Orders 09 and 14).

In that case, however, the parties reached a settlement. AR at 708. In accepting the settlement, the WUTC noted “‘that the testimony and trending data offered in support of the proposed rate increase for 2014 are substantially less precise than we would require in a fully-litigated rate case.’” AR at 708 (quoting Avista Orders 09 and 14 at 27) (emphasis omitted). It made clear that it was “‘not endorsing the specific attrition methodologies, assumptions, or inputs’” but did endorse the view “‘that an attrition adjustment should not be limited to circumstances where the utility can demonstrate extreme financial distress.’” AR at 709 (quoting Avista Orders 09 and 14 at 30). The WUTC announced its intent “‘to clarify the conditions wherein attrition should be considered when setting rates,’” but it felt limited in its ability to do so by the settlement. AR at 709 (quoting Avista Orders 09 and 14 at 30).

The current case presented the WUTC with the opportunity to address attrition in a “fully-litigated rate case.” *7 48982-1-II

III. MULTIPARTY SETTLEMENT

Avista, Staff, PCU, NWIGU, and ICNU came to a multiparty settlement pursuant to WAC 480-07-730.4 As a result of the settlement, Avista reduced its requested rate increases from $33.2 million to $17 million for electric and from $12 million to $11.3 million for gas. The settling parties agreed that Avista should receive a 9.5 percent return on equity and a 7.29 percent rate of return. The parties did not come to any agreement regarding attrition.

IV. EVIDENTIARY HEARING

On October 5, 2014, the WUTC commenced an evidentiary hearing to address the issues not resolved by the settlement. The disputed issues included whether to apply an attrition adjustment and, if so, the proper methodology for an attrition study.

A. AVISTA TESTIMONY AND ATTRITION STUDY

Scott Morris, Avista’s Board Chairman, President, and Chief Executive Officer, testified that Avista required a rate increase because net plant and operating expenses continued to grow, while customer and sales growth remained stagnant. Kelly Norwood, Avista’s Vice President of State and Federal Regulation, testified that Avista had earned less than its authorized return on equity from 2008 until 2013. It earned 9.5 percent in 2013 and 9.9 percent in 2014, both near its authorized return of 9.8 percent.5

Based on Morris’s calculated needs for a rate increase and projected increases in plant investment and operating expenses, Avista presented an attrition study for both its electric and

4 WUTC regulations allow for “[a]n agreement of some, but not all, parties on one or more issues” to be “offered as their position in the proceeding along with the evidence that they believe supports it.” WAC 480-07-730(3). 5 Avista over-earned on its electric and under-earned on its natural gas in 2013 and 2014, averaging out to near its authorized rate of return.

48982-1-II natural gas operations. During its rebuttal, however, it abandoned its own attrition study. Instead, it adopted the one proposed by Staff with several changes to the methodology used to calculate historical operations and maintenance costs.

B. STAFF TESTIMONY AND ATTRITION STUDY

Staff presented testimony of WUTC employee Chris R. McGuire. AR at 3808. McGuire explained that:

The [WUTC’s] standard ratemaking practice requires companies filing for revised rates to start with an historical test year, and typically with an average of monthly averages rate base balance. The [WUTC] also allows pro forma adjustments to rate base and expenses that often extend beyond the end of the test year. Such pro forma adjustments are what characterize Washington as a “hybrid” or “modified” historical test year state.

AR at 3819 (footnotes omitted). He also discussed the various requirements that go into what specific pro forma plant adjustments the WUTC considers in setting rates using this method.

McGuire testified that, to be included in pro forma plant adjustments under the traditional modified historical test year analysis, a given pro forma plant must: “1. Be in service as of June 30, 2015; 2. Meet a reasonable definition of ‘major’; 3. Demonstrate ‘quantifiable’ benefits to ratepayers in Washington State; and 4. Include only costs that were prudently incurred.” AR at 3822. He testified that 14 pro forma plant adjustments met these criteria.6 AR at 3823.

McGuire also testified as to when the WUTC should allow a utility to receive an attrition adjustment. He noted that:

[T]he [WUTC] has historically allowed attrition where extraordinary circumstances beyond the regulated company’s control will cause the relationship between revenues, expenses, and rate base to change in a manner that impedes the company’s opportunity to earn a fair return. The [WUTC] historically has provided attrition allowances to companies during times of rapid growth in utility plant.

6 Staff witness David Gomez detailed each of the 14 “major capital additions.” AR at 3976-4021. In its final order, the WUTC accepted his pro forma adjustment calculations to rate base before applying its attrition adjustment.

48982-1-II AR at 3824 (footnote omitted). Applying these criteria to Avista’s situation, he testified that “Avista’s low load growth is well-documented and its net plant has increased in recent years.” AR at 3824. However, he found that “the attrition evidence presented by [Avista] [did] not conclusively demonstrate that the circumstances currently facing [Avista] [were] sufficient to warrant an attrition allowance.” AR at 3824. Despite this conclusion, his attrition analysis “show[ed] that [Avista] [would] likely under-recover in the rate year at rates established using only a modified historical test period approach.” AR at 3824.

McGuire further testified that “rates calculated using a modified historical test year approach [would] likely be insufficient to provide [Avista] with a fair opportunity to earn the Settlement rate of return.”7 AR at 3813. McGuire prepared two exhibits representing Staff’s attrition study.

In calculating Avista’s revenue requirements, Staff’s attrition study “estimate[d] the revenue required to provide Avista with a fair opportunity to earn the settlement rate of return in 2016.” AR at 3817. Because it determined that the revenue requirement calculated with a modified historical test year approach would yield insufficient revenue, Staff recommended that the WUTC approve an attrition adjustment to Avista’s rates. However, McGuire also noted his concerns with Avista’s rapidly increasing investment in distribution plant and accused Avista of failing to provide a substantive explanation or any evidence justifying the rapid expansion.

McGuire summarized Staff’s findings using both a modified historical test period approach and the results of Staff’s attrition study. Staff’s analyses concluded that the modified historical test year approach would yield a rate short of Avista’s attrition-adjusted revenue requirement.

7 The settlement set Avista’s rate of return on investment at 7.29 percent.

48982-1-II Accordingly, McGuire recommended a $14.5 million attrition allowance for electric operations and a $6.7 million attrition allowance for natural gas operations.

To perform his attrition study and calculate Avista’s electric rate base, expenses, and other non-retail revenue in 2016, McGuire “relied upon historical, normalized data . . . to estimate rates of growth for 1) net plant after deferred income taxes; 2) operating expense; 3) total depreciation/amortization; 4) taxes other than income; and 5) other non-retail revenue.” AR at 3845. Once he had established the growth rates, he applied them to the 2014 levels of each category. McGuire calculated an escalation factor of 7.83 percent and applied that rate to each category of Avista’s 2014 rate base. McGuire thus calculated Avista’s projected 2016 rate base by applying this escalation factor to its 2014 rate base and applying various other factors and adjustments. McGuire used the same process to perform his attrition study for Avista’s natural gas rates, with an escalation factor of 16.86 percent.

On cross-examination by ICNU counsel, McGuire stated that he was “not testifying to the used and useful nature of any specific plant beyond July of 2015.” 5 WUTC Transcripts at 457. He further clarified that “an attrition allowance is an undistributed increase in revenue” and that the attrition study was “not any acceptance of some specific plant addition in the future.” 5 WUTC Transcripts at 457. McGuire also stated that his attrition study did not purport to make “any assessment of whether or not any investment is prudent or imprudent or will or will not be used and useful.” 5 WUTC Transcripts at 445. Rather, he characterized the attrition allowance as “an undistributed increase in revenue not associated with any specific plant” and stated that he was “not making assessments of individual plant in an attrition study.” 5 WUTC Transcripts at 445. *11 48982-1-II

C. OTHER TESTIMONY

ICNU presented testimony from Bradley Mullins, a consultant representing industrial customers. He testified that if “the [WUTC] were to disregard the pro forma cross check study, and instead establish revenue requirement solely on [Avista’s] proposed trending methodology,[8] it would allow [Avista] to earn a return on capital expenditures that have not been determined to be prudent and in the public interest.” AR at 6457. He also warned that “[i]f a trending analysis is used . . . the undistributed increase in rate base resulting from the reliance on historical trends would not represent known and measureable capital items, nor could it be determined whether the increase was the result of capital that is used and useful.” AR at 6457.

D. POST-HEARING BRIEFS

Avista, The Energy Project, NWIGU, ICNU, Staff, and PCU all filed post-hearing briefs arguing their various positions.

Avista requested an increase of $3.6 million for its electric rates and an increase of $10 million for its gas rates, predicated on an attrition analysis.

ICNU argued in its brief that “Avista’s attrition adjustment proposal falls well short of meeting” the “known and measurable” standard. AR at 451. It further cited case law for the proposition that “the unknown and unmeasurable future capital expenditures represented by [Avista’s] proposed attrition adjustment do not satisfy the statutory ‘used and useful’ requirement.” AR at 452.

Staff described attrition as “an extraordinary mechanism used only when extraordinary future circumstances can be demonstrated to negatively impact a utility’s financial integrity during

8 Mullins uses the terms “trending methodology” or “trending analysis” interchangeably with “attrition methodology” or “attrition analysis” throughout his testimony. See AR at 6452-53; 6457-58.

48982-1-II the rate effective year.” AR at 508. Given that Avista had earned at or above its approved rate of return in 2013 and 2014, Staff argued that it had “a reasonable opportunity to earn its allowed rate of return—even without an attrition allowance.” AR at 509.

Staff considered the WUTC’s test: “whether the company is facing extraordinary circumstances beyond its control.” AR at 510. Though Staff suggested that Avista had likely not shown that attrition should apply, it did present an attrition study. Staff described its attrition study as follows:

It is built upon historical data that is used to project nominal amounts of both net plant and expenses for the rate effective period. Mr. McGuire’s attrition proposal is not and should not be construed to represent a pro-forma examination of specific future capital expenditures. Rather, it is based on his judgment and analysis of [Avista’s] historical trends, including forecasted capital spending not otherwise accounted for in Staff’s case. Importantly, he did not examine the likelihood of Avista’s specific forecasted capital projects actually going into service in the rate year. Nor does Avista’s recent earnings history affect the results of Staff’s attrition study given that Mr. McGuire’s attrition analysis is forward looking. Mr. McGuire also does not address whether Avista can control its forecasted costs during the rate year.

AR at 511-12 (footnotes omitted).

In determining whether Avista should receive an attrition adjustment, Staff pointed out that Avista’s witnesses did “not explain the actual cause of why distribution plant expenditures are growing so quickly.” AR at 516. Though Avista demonstrated “a narrative of its capital budget,” it did not offer “a clear explanation of an extraordinary cause.” AR at 516.

PCU urged the WUTC to forego an attrition adjustment and set Avista’s rates using the more established modified historical test year methodology. It argued that Avista’s and Staff’s attrition studies were both flawed and failed to demonstrate that Avista was experiencing extraordinary events warranting a shift in traditional ratemaking. PCU warned that an attrition adjustment “cannot be a known and measurable change” because it is based on historic trends and *13 48982-1-II projected escalations, but it acknowledged that attrition is an appropriate response to a “‘trend of under-earning due to circumstances beyond the [c]ompany’s ability to control.’” AR at 371, 373 (quoting PSE Order 08 at 180).

Within its criticism of attrition’s inherent unreliability, PCU argued that “[w]ith either Avista’s or Staff’s attrition analysis, the [WUTC] would be required to approve capital investments that have not been demonstrated to be used and useful and trends that are not known and measurable.” AR at 378.

V. FINAL AGENCY ORDER

On January 6, 2016, the WUTC issued Order 05—its final order in the case. It issued findings of fact and conclusions of law.

A. WUTC ATTRITION ANALYSIS

i. Historical Treatment of Attrition

Summarizing its historical treatment of attrition in ratemaking, the WUTC observed that “the two most common sources of earnings attrition in Washington are abnormal or excessive inflation and exceptional and prolonged levels of plant additions.” AR at 711.

The WUTC declared that, unlike the previous two Avista ratemaking decisions in 2012 and 2014, “it is clear there is no agreement on the extent to which Avista suffers from attrition in either its electric or gas operations, nor is there consensus between [Avista] and Staff on the exact method for determining the extent of any reasonable attrition adjustment.” AR at 712. Therefore, it announced: “we must conduct a closer examination of the evidentiary record in determining whether and how to authorize an attrition adjustment.” AR at 712.

The WUTC determined that it must resolve three primary issues with regard to attrition in this case: “1) the appropriate criteria for determining whether an attrition adjustment is warranted; *14 48982-1-II 2) the appropriate methodology for an attrition study; and 3) whether Avista has met its burden of proof to justify granting an attrition adjustment for both electric and natural gas rates.” AR at 725.

ii. New Attrition Test

In “the early attrition cases,” the WUTC acknowledged that it “found extraordinary circumstances that supported the use of attrition in periods of high inflation and extraordinary levels of investment in production plant, among other criteria.” AR at 725. It acknowledged that no such circumstances were present in this case. However, it found that “Avista is making increased capital investments in non-revenue generating plant (primarily on the distribution system) in an environment of low load growth.” AR at 725. The WUTC characterized Avista’s circumstances as the “new normal.” AR at 725.

In considering its recent statements about attrition adjustments, the WUTC stated it had “not established a different standard or criteria for attrition adjustments in more recent cases,” but had indicated “that ‘an attrition adjustment should not be limited to circumstances where the utility can demonstrate extreme financial distress.’” AR at 726 (quoting Avista Orders 09 and 14 at 30).

The WUTC decided that “it is not necessary to require a finding of extraordinary circumstances to justify granting an attrition adjustment.” AR at 726. Rather, it concluded that:

An attrition adjustment is yet another tool in our regulatory “toolbox” for utility ratemaking. However, we do require that utilities requesting an attrition adjustment demonstrate that the cause of the mismatch between revenues, rate base and expenses is not within the utility’s control. Without such a standard, a utility could plan for a level of expenditures that would exceed revenues and rate base recovery, creating the need for an attrition adjustment.

AR at 726. The new test announced by the WUTC required only that the utility seeking an attrition adjustment demonstrate that the attrition it was experiencing be outside its control. *15 48982-1-II

iii. Attrition Study Methodology

The WUTC next articulated its perspective on the appropriate methodology for an attrition study. It found that Staff’s approach, with some of Avista’s modifications, “provide[d] the most appropriate methodology in this docket for supporting an attrition adjustment.” AR at 726. It concluded that the proper methodology “begins with development of a modified historical test year with pro forma plant additions, even subsequent to a test year. An attrition study is based on the resulting projected earnings and revenue requirements, and the attrition adjustment is added only if the study shows a mismatch of earnings and expenditures.” AR at 726.

The WUTC made two adjustments to Staff’s attrition study. First, it adopted Avista’s 2007-2014 time period, rather than Staff’s 2009-2014 period to calculate growth trends. Second, it rejected the operating and maintenance (O&M) expense growth rates provided by both Avista and Staff and opted for one between the two.

iv. Application of New Attrition Test

Finally, the WUTC set about determining whether Avista had met its burden of proof to justify an attrition adjustment: to demonstrate persuasively that the attrition occurring was outside of its control. In applying its new attrition methodology to Avista, the WUTC decided it was critical to determine whether “it [was] necessary for Avista . . . to demonstrate that its need to invest in non-revenue generating plant, particularly distribution plant, [was] so necessary and immediate as to be beyond its control.” AR at 727-28. *16 48982-1-II

In an effort to address concerns about the “Averch-Johnson effect,”9 the WUTC declared that, although it no longer required a showing of extraordinary circumstances to justify attrition, it still required a persuasive demonstration “that the attrition occurring is outside of [the utility’s] control.” AR at 728.

In applying its new methodology to determine whether Avista should receive an attrition adjustment for its natural gas services, the WUTC noted that Avista had “reasonably demonstrated that it is making significant investments in non-revenue generating plant for the purposes of safety and reliability.” AR at 729. It found that Avista had “established that the need for its capital investments in natural gas operations are beyond its control.” AR at 730. Accordingly, it “authorize[d] an attrition adjustment in accordance with the methodology advocated by Staff” with some exceptions. AR at 730. The WUTC granted an attrition adjustment of approximately $6.8 million for Avista’s natural gas operations, resulting in an overall increase in revenue requirement of $10.8 million.

In applying the new attrition methodology to Avista’s electrical services, the WUTC found that the evidence did “not convince [it] that Avista’s projected electric distribution investments [we]re entirely outside of its control, or [were] required for the safe and efficient operation of its system.” AR at 732. Because there was “some, but not complete, evidence” that the attrition was

9 ICNU witness Bradley Mullins described this effect as follows:

It has been widely documented that utilities subject to rate of return regulation have an incentive to over-invest in capital in order to increase earnings. This phenomenon is often referred to as the Averch-Johnson Effect, based on the economists who first developed the model to describe the phenomenon, and has a real and significant impact on how utility operations are managed.

AR at 6454 (citing Harvey Averch & Leland L. Johnson, Behavior of the Firm Under Regulatory Constraint, 52 AM. ECON. REV. 996, 1052 (1962).

48982-1-II outside Avista’s control, the WUTC decided that it “retain[ed] broad discretion to consider other factors, such as [Avista’s] intent to file another rate case within the next year, and the analysis under Hope,[10] Bluefield,[11] and Permian Basin.”12 AR at 733. The WUTC announced that it believed it could “exercise broad discretion to consider such seminal cases using [its] informed judgment in deciding whether or not an attrition adjustment is warranted given the specific facts and circumstances in a rate case.” AR at 733.

The WUTC noted that, if it were to reject the attrition adjustment for electric revenue, Staff’s modified historical test year analysis would reduce Avista’s electric revenue requirement by more than $20 million. Although PCU and ICNU recommended higher reductions than that, the WUTC determined that it could not “reasonably conclude such an end result would be appropriate under the standards in Hope and Bluefield.” AR at 734. It announced that its ratemaking responsibility “turns not on the particular rate making methodology it selects, i.e., modified historical test year or attrition, but on its outcome, or ‘end results.’” AR at 734 (citing Hope, 320 U.S. at 603). The WUTC then quoted the Washington Supreme Court’s analysis of Hope and Permian Basin, to conclude that “‘within a fairly broad range, regulatory agencies exercise substantial discretion in selecting the appropriate rate making methodology.’” AR at 735 (quoting POWER 85, 104 Wn.2d at 812).

The WUTC concluded that “[a] drastic rate reduction, such as proposed by parties that urge us to reject an attrition adjustment, would run afoul” of the ratemaking principles discussed in

10 320 U.S. at 603. 11 262 U.S. at 692. 12 In re Permian Basin Area Rate Cases, 390 U.S. 747, 791-92, 88 S. Ct. 1344, 20 L. Ed. 2d 312 (1968).

48982-1-II POWER 85. AR at 735. Based on its interpretation of POWER 85 and its analysis of Hope, Bluefield, and Permian Basin, the WUTC granted Avista an attrition adjustment for its electric operations using Staff’s attrition analysis, with two modifications. First, because Avista could not show that its proposed elevated investments in distribution plant were beyond its control, the WUTC removed all escalation of distribution plant in calculating Avista’s rate base. Second, it rejected both Avista’s and Staff’s calculation of the O&M escalation rate, instead calculating the escalation rate itself.

The WUTC applied an attrition adjustment of $28.3 million and concluded that Avista’s electric services revenue requirement should be reduced by approximately $8.1 million. Despite having granted Avista the attrition adjustment, the WUTC expressed frustration “about continuing to authorize recovery for these significant capital investments, absent a complete demonstration by [Avista] of quantifiable benefits to ratepayers.” AR at 737.

B. FINDINGS OF FACT AND CONCLUSIONS OF LAW

The WUTC found that “[t]he evidentiary record support[ed] a finding that Avista [would] experience attrition in its electric and natural gas operations over the rate effective year” and that “[a]bsent an attrition adjustment, [Avista] may not have an opportunity to achieve earnings on electric operations at or near authorized levels.” AR at 773.

The WUTC concluded that Avista had met its burden to demonstrate that its gas rates were insufficient to yield reasonable compensation and required an upward adjustment, but that it failed to meet its burden to show the same for its electric rates. To set new rates, the WUTC “accept[ed] and modif[ied] Staff’s attrition methodology for the purposes of setting rates for Avista’s natural gas operations as reasonable.” AR at 776. It also accepted Staff’s attrition methodology for setting electric rates, but “remove[d] any escalation of projected capital investments for distribution plant” *19 48982-1-II because distribution plant costs had not been demonstrated “as necessary or beyond [Avista’s] control.” AR at 776. It also modified the electric O&M expense escalation rate.

Avista filed tariff sheets reflecting the changes in rates. The new rates went into effect on January 11, 2016. 13

PCU petitioned the Thurston County Superior Court for judicial review of the WUTC’s decision and then applied for direct review by this court. The Superior Court certified the case to us for direct review pursuant to RCW 34.05.518.

ANALYSIS

I. WAIVER

PCU assigns error to the WUTC’s use of an attrition adjustment in determining Avista’s rates because it relied on estimates that were not associated with any particular “used and useful” Avista plant. The WUTC contends that this challenge is untimely because PCU never made this argument before the WUTC and thus may not raise it on appeal. Because ICNU raised the issue before the agency, we consider its merits.

A. LEGAL PRINCIPLES

The Washington Administrative Procedure Act (APA) provides that “[i]ssues not raised before the agency may not be raised on appeal” unless a specific exception applies. RCW 34.05.554(1). “In order for an issue to be properly raised before an administrative agency, there must be more than simply a hint or a slight reference to the issue in the record.” Citizens for Mount

13 PCU alleges that the WUTC erred in its calculations of the final attrition adjustment based on confusion relating to Avista’s updated power supply costs. The WUTC denied PCU’s and ICNU’s motions for clarification of its order and Staff’s motion for reconsideration, finding that the electric rates set by Order 05 were a fair, just, reasonable, and sufficient end result based on substantial evidence. Because we resolve the case on other grounds, we do not reach the alleged computational errors and do not discuss them further.

48982-1-II Vernon v. City of Mount Vernon, 133 Wn.2d 861, 869, 947 P.2d 1208 (1997). Washington law “require[s] issues to be first raised at the administrative level and encourage[s] parties to fully participate in the administrative process.” Citizens for Mount Vernon, 133 Wn.2d at 869.

Neither party addresses whether a party waives an argument when a different party raised the issue before the agency.

B. PCU’S TREATMENT OF THE ISSUE

The WUTC argues that PCU, for the first time on appeal, argues that attrition adjustments are never lawful because “they incorporate the value of rate base assets that are not ‘used and useful.’” Br. of Resp’t at 27.

PCU identifies one place in the record where it argued that an attrition adjustment would violate RCW 80.04.250’s “used and useful” language. In PCU’s post-hearing brief, it argued that “[w]ith either Avista’s or Staff’s attrition analysis, the [WUTC] would be required to approve capital investments that have not been demonstrated to be used and useful and trends that are not known and measureable.” AR at 378 (citing 5 WUTC Transcripts 612). Its argument was a single-sentence conclusory paragraph and its brief did not discuss the issue further. It cited ICNU witness Mullins’s testimony on cross-examination that the attrition study would require reliance on capital that was not used and useful. PCU did not cite any law, argue that this made the adjustment illegal, or explain the significance of the “used and useful” terminology. PCU’s treatment of the issue before the WUTC was one sentence in a fifty-nine page brief, in an administrative record spanning thousands of pages.

In King County v. Washington State Boundary Review Board, the county pointed to the ordinance at issue in the administrative record and a memorandum from another party arguing the *21 48982-1-II other side of the issue. 122 Wn.2d 648, 670, 860 P.2d 1024 (1993). The court held that these factors were insufficient to preserve the issue for judicial review. King County, 122 Wn.2d at 670.

In this case, PCU argued that the attrition adjustment considered investments that were not “used and useful” in its post-hearing brief in a single sentence. PCU’s treatment of the issue before the WUTC was exactly the type of “hint” or “slight reference” to the issue that is insufficient to preserve it on appeal. If PCU had been the only party raising this issue, it would have waived its “used and useful” argument.

C. ICNU’S TREATMENT OF THE ISSUE

PCU also points out where ICNU raised the “used and useful” issue before the WUTC. ICNU argued in its post-hearing brief that “according to the Supreme Court of Washington in POWER I,[14] the unknown and unmeasurable future capital expenditures represented by [Avista’s] proposed attrition adjustment do not satisfy the statutory ‘used and useful’ requirement.” AR at 452. ICNU argued that “the used and useful statute prohibits [Avista’s] proposal to incorporate indeterminate future capital amounts into tariff charges.” AR at 452.

We have not addressed whether a party must raise an argument before the agency when another party has raised the issue during the administrative proceedings. Federal courts considering the issue, however, have decided that the exhaustion requirement should be excused where the agency “‘has had an opportunity to consider the identical issues . . . but which were raised by other parties.’” Nat. Res. Def. Council, Inc. v. U.S. Envtl. Prot. Agency, 824 F.2d 1146, 1151 (D.C. Cir. 1987) (quoting Buckeye Cablevision, Inc. v. United States, 438 F.2d 948, 951 (6th

14 People’s Org. for Wash. Energy Res. v. Utils. & Transp. Comm’n, 101 Wn.2d 425, 430, 679 P.2d 922 (1984).

48982-1-II Cir. 1971)). The Washington Supreme Court has cited the D.C. Circuit for the policies underlying the waiver doctrine, noting that the rule furthers the purposes of:

“(1) discouraging the frequent and deliberate flouting of administrative processes; (2) protecting agency autonomy by allowing an agency the first opportunity to apply its expertise, exercise its discretion, and correct its errors; (3) aiding judicial review by promoting the development of facts during the administrative proceeding; and (4) promoting judicial economy by reducing duplication, and perhaps even obviating judicial involvement.”

King County, 122 Wn.2d at 669 (quoting Fertilizer Inst. v. U.S. Envtl. Prot. Agency, 935 F.2d 1303, 1312-13 (D.C. Cir. 1991).

These policies are all supported by allowing another party to preserve an issue. Further, the D.C. Circuit, where those policies come from, expressly allowed this practice so long as the agency “had an opportunity to consider” the issue. Nat. Res. Def. Council, 824 F.2d at 1151. Because ICNU placed the issue before the WUTC, we consider PCU’s statutory “used and useful” argument.

II. USED AND USEFUL

PCU contends that the WUTC “exceeded its statutory authority by setting rates for Avista’s utility services that included amounts for utility plant that were not ‘used and useful’” as required by statute. Br. of Appellant at 20. It argues that the WUTC considered “projected amounts of potential future utility plant” that was “not associated with actual utility plant that could be used to provide utility service, and is thus not ‘used and useful.’” Br. of Appellant at 23-24.

The WUTC responds that “rate base assets that are valued through an attrition adjustment are ‘used and useful’ within the meaning of RCW 80.04.250.” Br. of Resp’t at 30. It argues that the attrition adjustment derives asset values “through a process that predicts, with a high degree of certainty, capital investments that will improve or maintain the utility’s system” during the rate-*23 48982-1-II effective period. Br. of Resp’t at 30. Therefore, it claims that it “reasonably assumes that the associated investments will benefit current ratepayers.” Br. of Resp’t at 30.

RCW 80.04.250 gives the WUTC authority to “determine the fair value for rate making purposes of the property of any public service company used and useful for service in this state.” However, the WUTC exceeds its authority if it calculates a utility’s rate base including property that is not “used and useful” for service in Washington. People’s Org. for Wash. Energy Res. v. Utils. & Transp. Comm’n, 101 Wn.2d 425, 434, 679 P.2d 922 (1984) (POWER 84). We conclude that the WUTC’s use of attrition to calculate rate base violates the “used and useful” principle of

RCW 80.04.250.

A. LEGAL PRINCIPLES

We review the WUTC’s final orders under the APA, chapter 34.05 RCW. PacifiCorp v. Wash. Utils. & Transp. Comm’n, 194 Wn. App. 571, 586, 376 P.3d 389 (2016). “The party asserting the invalidity of the [WUTC’s] action has the burden of demonstrating invalidity.” PacifiCorp, 194 Wn. App. at 586.

We review the WUTC’s findings for substantial evidence supporting the finding. PacifiCorp, 194 Wn. App. at 586. Substantial evidence is “‘evidence sufficient to persuade a fairminded person of their truth.’” City of Vancouver v. Pub. Emp’t Relations Comm’n, 180 Wn. App. 333, 347, 325 P.3d 213 (2014) (quoting City of Vancouver v. Pub. Emp’t Relations Comm’n, 107 Wn. App. 694, 703, 33 P.3d 74 (2001))). The substantial evidence standard is highly deferential and unchallenged findings are verities on appeal. PacifiCorp, 194 Wn. App. at 586-87.

We will overturn an agency order if the order “is outside the statutory authority or jurisdiction of the agency conferred by any provision of law,” “is not supported by evidence that is substantial when viewed in light of the whole record before the court,” or is “arbitrary or *24 48982-1-II capricious.” RCW 34.05.570(3)(b), (e), & (i). We will also reverse if the agency has “erroneously interpreted or applied the law.” RCW 34.05.570(3)(d).

The legislature has tasked the WUTC to “[r]egulate in the public interest, as provided by the public service laws, the rates, services, facilities, and practices of all persons engaging within this state in the business of supplying any utility service or commodity to the public for compensation.” RCW 80.01.040(3). Among the duties the legislature has delegated to the WUTC is that to set rates for utilities, subject to appropriate judicial review. POWER 85, 104 Wn.2d at 808. “The [WUTC] must set ‘just, fair, reasonable and sufficient’ rates.” PacifiCorp, 194 Wn. App. at 587 (quoting RCW 80.28.010). The “function of rate-making is legislative in character, and the judicial branch is not empowered to undertake the job of fixing rates.” PacifiCorp, 194 Wn. App. at 587. The utility has the burden of proof for increasing its rates. PacifiCorp, 194 Wn. App. at 587.

The WUTC’s duties require it “to not only assure fair prices and services to customers, but also to assure that regulated utilities earn enough to remain in business.” POWER 85, 104 Wn.2d at 808. “A utility is not permitted to recover every expense in its rate structure; the WUTC ‘has the power to review operating expenses incurred by a utility and to disallow those which were not prudently incurred.’” Willman v. Wash. Utils. & Transp. Comm’n, 122 Wn. App. 194, 204, 93 P.3d 909 (2004) (quoting POWER 85, 104 Wn.2d at 810). However, the WUTC “must ‘assure that regulated utilities earn enough to remain in business.’” PacifiCorp, 194 Wn. App. at 588 (quoting Willman, 122 Wn. App. at 204).

“‘The [WUTC] has broad generalized powers in rate-setting matters.’” PacifiCorp, 194 Wn. App. at 588 (quoting US W. Commc’ns, Inc. v. Wash. Utils. & Transp. Comm’n, 134 Wn.2d 48, 56, 949 P.2d 1321 (1997)). Courts “are not at liberty to substitute their judgment for that of *25 48982-1-II the [WUTC] in rate cases” and “within a fairly broad range, regulatory agencies exercise substantial discretion in selecting the appropriate rate-making methodology.” US W. Commc’ns, Inc., 134 Wn.2d at 56.

We give “substantial deference to a regulatory agency’s judgment about how best to serve the public interest.” Att’y Gen.’s Office, Pub. Counsel Section v. Utils. & Transp. Comm’n, 128 Wn. App. 818, 824, 116 P.3d 1064 (2005). “An agency’s decision cannot be set aside absent a clear showing of abuse of discretion.” PacifiCorp, 194 Wn. App. at 588. We do not weigh evidence or judge witness credibility, and we defer to the WUTC’s discretion in weighing testimony of experts. PacifiCorp, 194 Wn. App. at 588-89.

However, statutory interpretation is a question of law we review de novo. Williams v. Tilaye, 174 Wn.2d 57, 61, 272 P.3d 235 (2012). “Where the language of a statute is plain, free from ambiguity, and devoid of uncertainty, there is no room for construction because the meaning will be discovered from the wording of the statute itself.” POWER 84, 101 Wn.2d at 429-30.

B. POWER CASES

PCU argues that the WUTC acted beyond the scope of its authority when it used an attrition adjustment to calculate Avista’s utility rates. We agree that WUTC’s use of an attrition adjustment to calculate Avista’s rate base violated RCW 80.04.250.

When setting rates, the WUTC has the power to “ascertain and determine the fair value for rate making purposes of the property of any public service company used and useful for service in this state and shall exercise such power whenever it deems such valuation or determination necessary or proper under any of the provisions of this title.” RCW 80.04.250(1) (emphasis added). When calculating a utility’s rate base for ratemaking, if the WUTC considers utility plant that “is neither employed for service nor capable of being put to use for service; . . . such a plant *26 48982-1-II is not ‘used and useful’ for service” and the WUTC exceeds its statutory authority. POWER 84, 101 Wn.2d at 430.

In POWER 84, the WUTC included construction work in progress (CWIP) in calculating the company’s rate base for determination of its electric rates. 101 Wn.2d at 427. It determined that “because of the magnitude of the company’s present and future construction programs, the company [would] not be able to finance its construction projects on reasonable terms without including a portion of CWIP in rate base.” POWER 84, 101 Wn.2d at 428. The Supreme Court analyzed the “used and useful” language of RCW 80.04.250(1) and held that an “uncompleted utility plant is neither employed for service nor capable of being put to use for service; therefore, such a plant is not ‘used and useful’ for service as required by RCW 80.04.250.” POWER 84, 101 Wn.2d at 430. Therefore, it concluded, “the [WUTC] exceeded its statutory authority by including CWIP in [the company’s] rate base.”15 POWER 84, 101 Wn.2d at 430.

The Court in POWER 84 further noted that RCW 80.04.250 contains no “indication that the [WUTC] possesses untrammeled discretion with respect to which property may be included in rate base for ratemaking purposes.” 101 Wn.2d at 434. Neither did it “find any indication in the statute that the financial condition of the utility is relevant in determining whether property may be included in rate base.” POWER 84, 101 Wn.2d at 434. In POWER 85, the Court also specified

15 In 1991, the legislature amended RCW 80.04.250 to expressly authorize the WUTC to include CWIP in calculating rate base. RCW 80.04.250; LAWS OF 1991, ch. 122, §2.

48982-1-II in dicta that the “property on which a public utility is entitled to earn a fair return is that which is used and useful for public service at the time the inquiry as to rates is made.” 16 104 Wn.2d at 815 (emphasis added).

In POWER 84, the WUTC had further erred by using rate base attributed to CWIP to calculate an attrition allowance for the company. 101 Wn.2d at 428. The court noted that a significant portion of the attrition allowance the WUTC had authorized was “directly attributable” to CWIP expenses. POWER 84, 101 Wn.2d at 428.

In POWER 85, the Supreme Court addressed another challenge to utility rates set by the WUTC. In that case, the WUTC had included costs associated with an abandoned power plant project in determining the company’s operating expenses. POWER 85, 104 Wn.2d at 814. POWER argued that the “used and useful” statute should be applied to operating expenses in the same way POWER 84 had applied it to calculation of rate base. POWER 85, 104 Wn.2d at 815. The Court held that RCW 80.04.250 “is purely a rate base statute and does not apply to operating expenses,” and so the costs of the abandoned power plants were not required to be “used and useful.” POWER 85, 104 Wn.2d at 815. Accordingly, insofar as the attrition adjustment is not a method of calculating rate base, it avoids the “used and useful” requirement of RCW 80.04.250.

C. ANALYSIS

PCU argues that, like the CWIP in POWER 84, the projections used to calculate Avista’s attrition adjustment “are not associated with any specific investment in rate base, and thus are not tied to utility plant employed for service or capable of being put to use for service.” Br. of

16 POWER 85 held that the WUTC may consider O&M expenses that are not “used and useful” when calculating a company’s operating expenses because RCW 80.04.250 “is purely a rate base statute.” 104 Wn.2d at 815. Its statement of the temporal element in RCW 80.04.250 therefore had nothing to do with its ultimate holding.

48982-1-II Appellant at 28. It contends that if “actual, unfinished plant under construction fails to meet the used and useful test, then mere projections of potential future capital investment also fail.” Br. of Appellant at 28.

The WUTC “exercise[s] substantial discretion in selecting the appropriate ratemaking methodology.” US W. Commc’ns, 134 Wn.2d at 56. However, it does not have discretion to include rate base that is not “used and useful.” POWER 84, 101 Wn.2d at 430. Therefore, if the attrition study and attrition adjustment was a methodology for determining Avista’s used and useful rate base and O&M expenses, the WUTC did not exceed its authority by using it. However, if the attrition study considered rate base that was not “used and useful,” the WUTC exceeded its authority.

The attrition adjustment the WUTC applied in this case was based on the study performed by McGuire. In that study, McGuire calculated Avista’s electric rate base for 2016 by determining its average rate of growth from 2009 to 201417 and applying that rate to the 2014 rate base. He then arrived at a rate base more than one hundred million dollars higher than the 2014 rate base. He used an identical process to calculate Avista’s natural gas rate base, resulting in a predicted 2016 rate base nearly fifty million dollars higher than the 2014 rate. McGuire acknowledged that he was “not testifying to the used and useful nature of any specific plant beyond July of 2015.” 5 WUTC Transcripts at 457. He also made clear that his attrition study did not purport to make “any assessment of whether or not any investment is prudent or imprudent or will or will not be used and useful.” 5 WUTC Transcripts at 445.

The WUTC made modifications to McGuire’s attrition study in its final order. In the tables it attached to its final order, it did not include any of McGuire’s heightened attrition numbers in

17 The WUTC ultimately adjusted the time period to 2007–2014 as suggested by Avista.

48982-1-II rate base. Rather, it included only the uncontested rate base adjustments and numbers reflecting planned capital additions for 2014 and 2015, based on the pro forma calculations. Then, after calculating rate base among other factors, it applied an attrition allowance it reached based on McGuire’s attrition study, with the WUTC’s modifications, at the end.18

Although the WUTC’s final attrition analysis excluded all escalation of capital investment in distribution plant, McGuire’s attrition analysis applied escalation to four other types of plant and several other categories of rate base,19 none of which was excluded by the WUTC. Rather than assessing Avista’s used and useful 2014 rate base, this process projected Avista’s likely rate base in 2016 using historical trend data. As acknowledged by McGuire, these projections are not based on “assessments of individual plant,” but general projections of Avista’s likely rate base as a whole.

In POWER 84, specific, identifiable, but incomplete plant was not “used and useful” as the term is meant in RCW 80.04.250. 101 Wn.2d at 430. Here, the WUTC based its calculation of Avista’s rate base on projections rather than any specific identifiable plant. Regardless of their accuracy, use of projections in determining rate base ignores the requirement that the WUTC only consider property “used and useful for service” in Washington. Like the CWIP in POWER 84, the projected data in this case is not tied to any particular used and useful plant.

We clarify that, as the Supreme Court ruled in POWER 85, RCW 80.04.250 “is purely a rate base statute and does not apply to operating expenses.” 104 Wn.2d at 815. To the extent the WUTC relied on its attrition adjustment to account for increases in Avista’s O&M expenses, it did

18 It is unclear where the exact numbers the WUTC reached in its attrition study came from. The record indicates that it held two order conferences in January and February of 2016 where an expert explained the numbers, but nothing from those conferences is in the record. 19 These other types were intangible, production, transmission, and general plant.

48982-1-II not violate the statute. However, its reliance on projections to calculate Avista’s rate base was improper.

The WUTC urges us to view its attrition study as a method of calculating rate base that we must defer to. It asserts that the “asset values are derived through a process that predicts, with a high degree of certainty, capital investments that will improve or maintain the utility’s systems during the period in which the newly approved rates will remain in effect.” Br. of Resp’t at 30. The WUTC does “exercise substantial discretion in selecting the appropriate ratemaking methodology.” US W. Commc’ns, 134 Wn.2d at 56. However, it does not have “untrammeled discretion with respect to which property may be included in rate base for ratemaking purposes.” POWER 84, 101 Wn.2d at 434. Because its attrition adjustment accounted for rate base that has not been shown to be “used and useful” in Washington, the WUTC exceeded its statutory authority.

The WUTC quotes a law review article for the proposition that costs and investments are “used and useful if . . . there is a direct and immediate benefit to customers; traditionally, the investment is made in plant that is operational now or in a future test year or in the period during which the rates may reasonably be expected to be in effect.” Br. of Resp’t at 31-32 (quoting James J. Hoecker, “Used and Useful”: Autopsy of a Ratemaking Policy, 8 ENERGY L.J. 303, 312 (1987)). However, this fails to account for the Washington Supreme Court’s conclusion that “property on which a public utility is entitled to earn a fair return is that which is used and useful for public service at the time the inquiry as to rates is made.” POWER 85, 104 Wn.2d at 815.

Avista cites numerous out-of-state cases for the proposition that “whether utility property is ‘used and useful’ and therefore to be included in rate base, is a factual determination.” Br. of Intervenor-Resp’t at 21. This is directly at odds, however, with POWER 84, which addressed whether CWIP was “used and useful” as a matter of statutory construction. 101 Wn.2d at 429-30. *31 48982-1-II Avista also argues that we should defer to the WUTC’s reasonable interpretation of RCW 80.04.250. Br. of Intervenor-Resp’t at 24. Despite this principle of construction, however, the POWER 84 court held that RCW 80.04.250 is “plain, free from ambiguity, and devoid of uncertainty,” such that “there is no room for construction because the meaning will be discovered from the wording of the statute itself.” 101 Wn.2d at 429-30.

CONCLUSION

Because the projections of future rate base were not “used and useful” for service in Washington, we conclude that the WUTC may not base Avista’s rates on them. Accordingly, the WUTC erred in calculating Avista’s electric and natural gas rates. The WUTC order provided one lump sum attrition allowance without distinguishing what portion was for rate base and which was for O&M expenses or other considerations. We strike all portions of the attrition allowance attributable to Avista’s rate base and reverse and remand for the WUTC to recalculate Avista’s rates without relying on rate base that is not used and useful.20

Melnick, J. We concur:

Johanson, J.

Lee, A.C.J.

20 PCU also alleges that the WUTC arbitrarily and capriciously applied its new attrition test to Avista’s electric rates. Br. of Appellant at 31. Because the WUTC exceeded its statutory authority by using attrition to calculate Avista’s rate base, we do not reach this issue.